Barbara Barrett is confused.
The 58-year-old retired teacher and elementary school counselor, who lives outside of Spokane, Washington, is due for a mammogram. But Tuesday’s headlines about mammograms leading to overtreatment in a “third of breast cancer patients” have given her — and thousands of other women across the country — huge cause for concern.
“I have increased anxiety about what I would do if something were found during my mammogram,” she said. “And I’m definitely confused reading the new info. Is there a way to know whether a tumor found in a mammogram should be treated or left alone?”
The headlines are based on a Danish study and an accompanying commentary by the chief medical officer of the American Cancer Society published Monday in the Annals of Internal Medicine. Both the study and commentary point to one of mammography’s biggest flaws: Along with picking up aggressive cancers that can go on to kill, they can also pick up slow-growing tumors that may never amount to anything. These “early cancers” include DCIS (ductal carcinoma in situ) or stage 0 breast cancers that are sometimes, but not always, harmless.
Unfortunately, we’re not quite yet able to consistently differentiate between the early-stage cancers that kill and the cancers that simply creep along (not to mention creep you out). So some women diagnosed with breast cancer may get, say, a mastectomy rather than a lumpectomy. Or get a lumpectomy and radiation for their DCIS when hormonal therapy alone might suffice. Some may even get chemotherapy that they don’t really need because a mammogram has only pointed to cancer — not a crystal clear vision of what the future holds.
Early cancers don’t always advertise their intent. They’re merely diagnosed and treated and sometimes, yes, overtreated. And that’s a huge concern — for cancer patients, for researchers and oncologists and, perhaps most of all, for women like Barbara Barrett who are trying to make health decisions while being inundated by headlines like: “Third of breast cancer patients treated unnecessarily, study says,” and “Time to rethink mammograms, American Cancer Society top doc says.”
A married mother of five and grandmother of (almost) two, Barrett is at average risk for the disease. As such, she knows she’s supposed to get regular mammograms, as recommended by her doctor, who follows the latest American Cancer Society guidelines. What she doesn’t know, especially in light of all the confusing, conflicting information swirling around out there, is how valuable mammograms actually are.
Will they find an invasive cancer at an early stage, as intended? Or will they, as these headlines suggest, actually cause her more harm than good? And if they do find a cancer, what does she do?
“I have lived through the horrible process of breast cancer treatment with my best friend. I wouldn’t wish it on my worst enemy,” she said. “To think that people are subjected to the horrible effects of chemo, radiation and mastectomy for no reason is heartbreaking. But I don’t think there is any way I would take the risk of declining treatment. I hope Fred Hutch and other researchers can figure it out soon.”
We tapped experts at Fred Hutch and its clinical care and research partner Seattle Cancer Care Alliance, who are working to improve the breast cancer screening process. Read on for their thoughts on this latest study and the key questions it’s raised about the value of mammography.
“There’s clear evidence that mammography continues to save lives,” said Dr. Habib Rahbar, clinical director of Breast Imaging at SCCA and associate professor of radiology at the University of Washington. “It’s something we feel very strongly about. Routine screening will catch cancers at an earlier stage, and we know that catching them at an earlier stage — in general — improves a patient’s outcome and increases the likelihood of survival.”
Rahbar said he and his colleagues have seen with their own eyes that mammograms “catch cancers that matter and catch them early.” And there is an incredible amount of scientific data — much of it gleaned through the gold standard of research, randomized controlled trials — to back this up.
“Mammography is one of the most investigated modes of screening,” he said. “We do what we do because there are both anecdotes and real scientific data to support its use.”
But, he added, that “doesn’t mean we don’t continue to strive to make a better test and improve it.”
There are a couple of them, actually.
Mammography works, but it’s a blunt tool — and it’s also a little fickle. Sometimes it doesn’t work well enough — it may miss tumors hidden within dense breast tissue (high risk clinics and 3-D imaging have helped combat this). Other times, it works too well: identifying “cancers” that may never come to fruition.
“Yes, if you go into screening, there’s a chance that you’ll have a low-risk cancer diagnosed,” said Fred Hutch’s Dr. Ruth Etzioni, a biostatistician and national expert on cancer screening who helped created the ACS’s recent breast screening guidelines. “That is a possibility. But if you have a low-risk cancer diagnosed, a cancer that doesn’t need to be treated, that doesn’t mean screening doesn’t work for someone else. We know there’s such a thing as overdiagnosis, but it doesn’t mean screening doesn’t work.
“They’re different sides of the coin,” she said. “It’s a flaw, but it’s not a flaw in mammography. It’s a flaw in the endeavor to detect cancer early.”
Rahbar summoned the classic baby-with-the-bathwater analogy.
“Just because it’s clear we have overdiagnosis doesn’t mean we should throw out an effective tool that decreases breast cancer-related deaths,” he said.
Women may also want to keep in mind that breast cancer screening isn’t just about mammography. Other screening methods such as ultrasound and MRI were not factored into the Danish study but are often used as supplemental screening for women at high risk and for breast cancer survivors.
Nope, Etzioni said. The numbers are too high. A more trustworthy estimate would be on the order of 10 to 15 percent, she said.
“It’s very complicated, but the vast majority of published studies on this are wrong because they use flawed methodology,” said Etzioni, who’s spent years trying to debunk these types of inflated statistics.
Etzioni and Fred Hutch colleagues have done a number of studies, she said, showing that the “simple-minded approach used by this and other highly cited studies is provably wrong in most cases.”
Overtreatment definitely exists, she emphasized, but not to the extent claimed by the Danish researchers.
“We don’t truly know the exact extent of overdiagnosis,” she said. “But most published numbers like these are too high.”
Getting truly accurate data would mean researchers would have to follow women who’d been diagnosed with breast cancer but not treated for it, Rahbar said. And that’s too dangerous.
“We could solve this issue by watching enough cancers and figuring it out,” he said. “But obviously the stakes are too high. What’s absolutely clear, though, is the great majority of invasive breast cancers would kill if left untreated.”
Rahbar said that although it’s true some DCIS lesions might not need to be treated, there’s currently no reliable way to determine this. The good news, he said, is that “these lesions are the earliest form of breast cancer and are extremely curable.”
So what should a woman do if she gets a mammogram and is told she has DCIS?
“I would absolutely recommend that any patient diagnosed with DCIS seek treatment at a comprehensive cancer center that can provide a range of treatment options — including clinical trials — that best match the patient’s preferences and goals,” he said.
In the meantime, research is definitely moving us ahead.
Fred Hutch and SCCA, he said, are both “very actively involved” in studies aimed at providing a more tailored, nuanced and individualized approach to women diagnosed with breast cancer, pointing to the the 75-site ECOG ACRIN trial and others coming down the pipeline. The new COMET trial, Etzioni said, will randomize women with low-risk DCIS to standard care or surveillance with treatment only upon evidence of disease recurrence or progression. A new Digital Mammography DREAM Challenge involving Fred Hutch and SCCA researchers aims to fast-forward breast screening technology through machine learning.
Etzioni and Rahbar also stressed that while there may be back and forth about how many women may or may not be overtreated because of mammograms, there is consensus about what needs to be done moving forward.
“I think everyone in the breast oncology world is in total agreement that we need to apply more resources towards advanced molecular and genetic testing of tumors, as well as imaging profiling of tumors and even precancerous tissue, to determine who needs more aggressive therapies and who we can treat less aggressively,” said Rahbar.
“The data presented doesn’t change the general direction we need to go as breast cancer specialists — towards more personalized, individualized care. “
Diane Mapes is a staff writer at Fred Hutchinson Cancer Research Center. She has written extensively about health issues for NBC News, TODAY, CNN, MSN, Seattle Magazine and other publications. A breast cancer survivor, she blogs at doublewhammied.com and tweets @double_whammied. Email her at email@example.com.